🏢 Australian National University

This paper examines the trade-offs between two popular diagonal Fisher Information Matrix (FIM) estimators in neural networks, deriving variance bounds and highlighting the importance of considering e…

This paper introduces a novel framework for classification with rejection by learning density ratios between data and idealized distributions, improving model robustness and accuracy.

PACE marries parameter-efficient fine-tuning with consistency regularization to significantly boost model generalization.

LAM3D: A novel framework uses point cloud data to boost single-image 3D mesh reconstruction accuracy, achieving state-of-the-art results in just 6 seconds.

Researchers devised a novel method to accelerate neural network training by guiding the optimization process toward a Simplex Equiangular Tight Frame, exploiting the Neural Collapse phenomenon to enha…

DreamSteerer enhances source image-conditioned editability in personalized diffusion models via a novel Editability Driven Score Distillation objective and mode shifting regularization, achieving sign…